ブランチ janitor/segclass-eucjp-literal-removal として、下記の変更を行
いました。
anthy_seg_class_name は Debug 出力としてのみ使っているので、
日本語を含まない anthy_seg_class_sym を使うことにしています。
diff --git a/anthy/segclass.h b/anthy/segclass.h
index b8a2724..b58113e 100644
--- a/anthy/segclass.h
+++ b/anthy/segclass.h
@@ -43,8 +43,5 @@ enum seg_class {
SEG_SIZE
};
-const char* anthy_seg_class_name(enum seg_class sc);
const char* anthy_seg_class_sym(enum seg_class sc);
-enum seg_class anthy_seg_class_by_name(const char *name);
-
#endif
diff --git a/src-splitter/lattice.c b/src-splitter/lattice.c
index 0387727..0d1cab2 100644
--- a/src-splitter/lattice.c
+++ b/src-splitter/lattice.c
@@ -193,7 +193,7 @@ calc_probability(int cc, struct feature_list *fl)
if (anthy_splitter_debug_flags() & SPLITTER_DEBUG_LN) {
anthy_feature_list_print(fl);
- printf(" cc=%d(%s), P=%f\n", cc, anthy_seg_class_name(cc), prob);
+ printf(" cc=%d(%s), P=%f\n", cc, anthy_seg_class_sym(cc), prob);
}
return prob;
}
diff --git a/src-splitter/metaword.c b/src-splitter/metaword.c
index 0491035..723a677 100644
--- a/src-splitter/metaword.c
+++ b/src-splitter/metaword.c
@@ -93,7 +93,7 @@ anthy_do_print_metaword(struct splitter_context *sc,
printf("*meta word type=%s(%d-%d):score=%d:seg_class=%s",
anthy_metaword_type_tab[mw->type].name,
mw->from, mw->len, mw->score,
- anthy_seg_class_name(mw->seg_class));
+ anthy_seg_class_sym(mw->seg_class));
print_metaword_features(mw->mw_features);
printf(":can_use=%d*\n", mw->can_use);
if (mw->wl) {
diff --git a/src-splitter/segclass.c b/src-splitter/segclass.c
index 6b440a4..f4dc02a 100644
--- a/src-splitter/segclass.c
+++ b/src-splitter/segclass.c
@@ -5,19 +5,24 @@
#include <anthy/segclass.h>
#include "wordborder.h"
-static struct {
- const char *name;
- const char *sym;
-} seg_class_tab[] = {
- {"文頭", "H"}, {"文末", "T"}, {"文節", "B"},
- {"接続語", "C"}, {"名詞+格助詞", "Nk"}, {"名詞+終端", "Ne"},
- {"動詞+付属語", "Vf"}, {"動詞+終端", "Ve"}, {"形容詞", "A"},
- {"形容動詞", "AJV"},
- {"連用修飾", "YM"}, {"連体修飾", "TM"},
- {"名詞", "N"}, {"名詞+付属語", "Nf"}, {"名詞+連用", "Ny"},
- {"動詞+連用", "Vy"},
- {"動詞+連体", "Vt"},
- {NULL, NULL}
+const char *seg_class_tab[] = {
+ "H" /* Head of sentence: Buntou */
+ "T" /* Tail of sentence: Bunmatsu */
+ "B" /* Segment: Bunsetsu */
+ "C" /* Setsuzokugo */
+ "Nk" /* Meishi+Kakujoshi */
+ "Ne" /* Meishi+Shuutan */
+ "Vf" /* Doushi+Fuzokugo */
+ "Ve" /* Doushi+Shuutan */
+ "A" /* Keiyoushi */
+ "AJV" /* Keiyoudoushi */
+ "YM" /* RenyouShuushoku */
+ "TM" /* RentaiShuushoku */
+ "N" /* Meishi */
+ "Nf" /* Meishi+Fuzokugo */
+ "Ny" /* Meishi+Renyou */
+ "Vy" /* Doushi+Renyou */
+ "Vt" /* Doushi+Rentai */
};
void
@@ -107,24 +112,7 @@ anthy_set_seg_class(struct word_list* wl)
wl->seg_class = seg_class;
}
-const char* anthy_seg_class_name(enum seg_class sc)
-{
- return seg_class_tab[sc].name;
-}
-
const char* anthy_seg_class_sym(enum seg_class sc)
{
- return seg_class_tab[sc].sym;
-}
-
-enum seg_class
-anthy_seg_class_by_name(const char *name)
-{
- int i;
- for (i = 0; seg_class_tab[i].name; i++) {
- if (!strcmp(seg_class_tab[i].name, name)) {
- return i;
- }
- }
- return SEG_BUNSETSU;
+ return seg_class_tab[sc];
}
diff --git a/src-splitter/wordlist.c b/src-splitter/wordlist.c
index 58c60bf..f677a3c 100644
--- a/src-splitter/wordlist.c
+++ b/src-splitter/wordlist.c
@@ -65,7 +65,7 @@ anthy_print_word_list(struct splitter_context *sc,
wl->part[PART_POSTFIX].len].c;
anthy_putxstr(&xs);
anthy_print_wtype(wl->part[PART_CORE].wt);
- printf(" %s%s\n", anthy_seg_class_name(wl->seg_class),
+ printf(" %s%s\n", anthy_seg_class_sym(wl->seg_class),
(wl->is_compound ? ",compound" : ""));
}
--
_______________________________________________
Anthy-dev mailing list
Anthy-dev@lists.sourceforge.jp
http://lists.sourceforge.jp/mailman/listinfo/anthy-dev
